Magnetic and Bathymetric Survey of the Suiyo Cross-Chain, Izu-Bonin Arc

Transcription

1 JAMSTEC Report of Research and Development, Volume 6, November 2007, Magnetic and Bathymetric Survey of the Suiyo Cross-Chain, Izu-Bonin Arc Toshiya Fujiwara 1, Osamu Ishizuka 2, Yoshihiko Tamura 1, Nobuo Geshi 2, Alison Shaw 3, Julie O'Leary 4, Hitoshi Tanaka 5 and Satomi Minamizawa 5 Abstract During the R/V Natsushima NT07-07 cruise, magnetic and bathymetric surveys of the Suiyo Cross-Chain were conducted. The newly collected bathymetric data fill gaps of swath coverage of previous cruises in shallow depths, therefore, a complete data set of swath bathymetry is obtained. The West Suiyo Volcanic Complex (WSVC) is magnetized in totality positive, and has more complex distribution of the magnetization intensities and/or directions. The Suiyo Volcano (SV) is strongly magnetized. The magnetization intensities increase from the WSVC through the Suiyo Volcanic Ridge to the SV, i.e. from back-arc to volcanic front. It may indicate the SV mainly consists of younger basaltic rocks, while the WSVC consists of more differentiated rocks and/or is older. The western flank of SV is magnetized in positive, in contrast, the eastern flank may be magnetized in negative. Otherwise, the SV is interpreted as the magnetization of eastward declination. The deflection of the magnetic declination can be explained by the clockwise rotation of the seamount together with the Philippine Sea Plate. In the vicinity of the caldera of the SV has low or reverse magnetization. Low magnetization is likely due to hydrothermal activity in the caldera. Keywords: Izu-Bonin Arc; Suiyo Seamount; bathymetry; magnetic anomaly. 1. Introduction The Suiyo Seamount is a volcano situated on line of the volcanic front in the southern Izu-Bonin Arc (Figure 1). The seamount is on the thinnest crust along the arc. The magma origin of the seamount is clearly distinguished in isotopic composition from that in the northern Izu-Bonin Arc [Ishizuka et al., 2007]. The West Suiyo Volcanic Complex (WSVC) is a volcano located at ~40 km west from the Suiyo Seamount. The detailed morphology of the WSVC was revealed by a bathymetric survey with the R/V Yokosuka (YK97-10). The WSVC is the only backarc seamount in the southern Izu-Bonin Arc. Comparison with these two volcanoes and volcanoes in the northern arc is essential to investigate spatial variations, across- and along-axis, of magma composition of this oceanic arc. Therefore, we carried out a research cruise (NT07-07: April-May 2007) with R/V Natsushima and ROV Hyper- Dolphin (PI: O. Ishizuka). The main objective of the cruise was geological observation and rock sampling on the Suiyo Cross-Chain (SCC). Hereafter, we term the Suiyo Volcano (SV) as the Suiyo Seamount, the Suiyo Volcanic Ridge (SVR) as the seamounts between the SV and the WSVC, and the Suiyo Cross-Chain (SCC) collectively (see Figure 6 for the location). During the cruise, magnetic and bathymetric surveys were also conducted. In this paper, we present the bathymetry and magnetic anomaly of the SCC. The bathymetric and magnetic study provides constraints on the crustal structure, bulk composition consisting of the seamounts, and eruption ages combined with geological, geochemical, and rock's age analyses. 2. Data Collection 2.1 Magnetic Survey The magnetic survey was conducted in nighttime during 4/25/2007-5/8/2007 and on ROV's maintenance day 1. Institute for Research on Earth Evolution (IFREE), Japan Agency for Marine-Earth Science and Technology (JAMSTEC) 2. Geological Survey of Japan, National Institute of Advanced Industrial Science and Technology (AIST) 3. Department of Geology and Geophysics, Woods Hole Oceanographic Institution 4. Department of Terrestrial Magnetism (DTM), Carnegie Institution of Washington 5. Marine Science Department, Nippon Marine Enterprises 31

2 Magnetic and Bathymetric Survey of the Suiyo Cross-Chain, Izu-Bonin Arc (5/3) during the NT07-07 cruise. Mean ship's speed during the survey was 8 kt (14.8 km/h). Main survey tracks are oriented to N-S, expanding 20 miles (37 km) centered at the SCC (Figure 2). Interval of the tracks is 1 mile. The track interval is very close compared with previous magnetic surveys [e.g. Yamazaki et al., 1991; Seama et al., 2005], thus obtaining high-resolution magnetic anomaly is promising. Some E-W oriented tracks are obtained to estimate cross over errors. Geomagnetic total force data were obtained by using a surface-towed cesium magnetometer G-882J (Geometrics) owned by Geological Survey of Japan (Figure 3(a)). The cable was attached to the nose of sensor towfish for high-speed towing (Figure 3(b)). The sensor was towed ~230 m behind the ship (Figure 3(c)). As for special coordination for the magnetic survey, a ship's capstan was fattened its diameter for retrieving the sensor towfish. The data logging system was settled in NO.1 Laboratory (Figures 3(d) and 4). Magnetic and GPS data were collected every 1 second using a MagLog software (The exception was 4/25's survey: every 0.1 seconds). Since the distance from the ship's GPS antenna to the stern is 53 m, thus the location of the magnetic sensor is ~285 m backward of the ship's Figure 1: Bathymetry of the Izu-Bonin Arc showing the location of the Suiyo Cross-Chain (SCC). The black square shows the survey area. SV: Suiyo Volcano, SVR: Suiyo Volcanic Ridge, WSVC: West Suiyo Volcanic Complex. 32

3 T. Fujiwara et al., Figure 2: Ship tracks of the R/V Natsushima where geophysical data are obtained. The illuminated image shows bathymetry. Figure 3: (a) G-882J cesium magnetometer system. The towfish has 1.4 m length and 20 kg weights. The cable is 15 mm diameter and 300 m length. (b) The cable is attached to the nose of towfish. (c) Retrieving the magnetometer towfish. (d) Data logging system settled in NO.1 Laboratory. The system is mainly composed of a personal computer, a junction box, and a USB serial adapter. 33

4 Magnetic and Bathymetric Survey of the Suiyo Cross-Chain, Izu-Bonin Arc position (Figure 4). The tow point was 6 m offset the starboard side from the GPS antenna. 2.2 Bathymetric Survey Simultaneously with the magnetic survey, bathymetric data were collected using a SeaBat 8160 multi-narrow beam echo-sounder system, which has frequency of 50 khz beams and a swath width of 150. The newly collected data fill gaps of swath coverage of previous cruises (YK97-10 and KR01-15) in shallow depths, in particular, at summits of seamounts. Therefore, a complete data set of swath bathymetry in the study area is obtained (Figures 5 and 6). 3. Observed Magnetic Anomaly The magnetic data were processed through the following sequence. The magnetic data were merged with SOJ data (navigation) taking into account the sensor position. Geomagnetic total force anomaly was calculated by subtracting the International Geomagnetic Reference Field (IGRF) 10th generation [IAGA, 2005] as the reference field. Diurnal correction is not applied, currently. The observed magnetic anomaly is shown in Figure 7. Large amplitudes of magnetic anomalies are observed at the SV. The western flank of SV is generally in positive anomaly. Positive peaks (~1050 nt in amplitude) are located at the northwestern portion and to the south of the SV. Local anomaly low is found in the south of the caldera over the western summit of the SV. The eastern flank of SV is in negative anomaly. The negative anomaly is extending north to south. Trough of the negative anomaly (~-350 nt in amplitude) is situated near the peak of the eastern SV. Above the WSVC, generally, positive anomaly is situated in the south, while negative anomaly appears in the north. The amplitude of the magnetic anomaly (~400 nt peak to trough) is smaller than that of the SV. As for details, the negative anomaly in the north of WSVC is divided into three troughs, and mid portion of the negative anomaly is rather high. The southwestern part of the WSVC is in rather low anomaly area. The SVR also shows a pattern that is positive anomaly in the south and negative anomaly in the north. The amplitude of the magnetic anomaly from the peak to trough is ~300 nt. 4. Magnetic Analysis As the preliminary analysis, magnetic anomaly due to a uniformly magnetized terrain is calculated to correct terrain effects on the observed anomaly and to evaluate magnetization intensity and direction. Top depth of the magnetized terrain follows bathymetry. Bottom depth of the model corresponds to be 10 km below the sea-surface. The terrain model is composed of a set of prism-shaped magnetic bod- Figure 4: Schematic figure of the magnetic survey in the NT07-07 cruise. 34

5 T. Fujiwara et al., Figure 5: Compiled swath bathymetry with 50 m contours based on data collected using a SeaBat (NT07-07), a HS-10 (YK97-10), and a SeaBeam 2112 (KR01-15). Bold lines with numbers show dive tracks of the ROV. Red and white stars indicate arriving and leaving points, respectively. Figure 6: Whale's-eye view of the SCC showing the West Suiyo Volcanic Complex (WSVC), the Suiyo Volcanic Ridge (SVR), and the Suiyo Volcano (SV). 35

6 Magnetic and Bathymetric Survey of the Suiyo Cross-Chain, Izu-Bonin Arc ies. The prisms are 0.5' (0.9 km 0.8 km) in horizontal extent near by observation points. They are 1.0' (1.9 km 1.6 km) far from the observation points (Figure 8). Magnetic prisms in size of 1000 km are placed in the further circumferential area to avoid edge effects. Average depth of the study area is given as the top depth of the prisms in the circum-area. The magnetization intensity is set to +5 A/m. The direction of magnetization in the source layer is assumed to be oriented parallel to a geocentric dipole field at the present latitude (declination 0, inclination 47 ). The ambient geomagnetic field is set to declination of -4 and inclination of 39 referring to IGRF. The calculation of magnetic anomaly due to the prism model is done by using a formulation of Bhattacharyya [1964] and Blakely [1996]. The calculated magnetic anomaly shows the case that the SCC has the simplest magnetization with uniform intensity and direction (Figure 9). In the figure, a typical anomaly pattern for a seamount located in the mid-latitude of northern hemisphere is visible. That is a dipole anomaly whose positive anomaly appears over the southern flank of a seamount and negative anomaly over the northern flank. 5. Discussion Compare the observed magnetic anomaly (Figure 7) with the calculated anomaly (Figure 9), the WSVC shows the anomaly pattern similar to the dipole anomaly in general. However, the pattern is not consistent with in detail. The result indicates that the WSVC is magnetized in totality positive, and has a more complex distribution of the magnetization intensities and/or directions. The amplitude in the observed anomaly is smaller than that of the calculated anomaly. The result suggests that bulk magnetized intensity of the WSVC is weaker than the assumed intensity of +5 A/m. Rock samples collected by the ROV dives during the cruise suggest that the WSVC consists of various types of rocks such as basalts, dacites, and rhyolites. Geological observation by the ROV also suggests that, probably, there are eruption age variations, resulting in polarities of remnant magnetization. Possible explanations for the resultant weak body magnetization are that the WSVC consists of mainly differentiated rocks which have weaker magnetization, and/or consists of older rocks which have weaker magnetization than younger rocks. The body magnetization may be an average of each magnetized layer to the depth or the horizontal direction. The each layer has remnant magnetization acquired when it has erupted. Because the seamount consists of volcanic rocks erupted in various ages, the volcanic rocks have erupted in a different magnetic epoch in history. Thereby obtained remnant magnetization of opposite magnetic polarity cancel Figure 7: Magnetic anomaly in the SCC. Contour interval is 50 nt. Thick black lines delineate bathymetry of 200 m contours. 36

7 T. Fujiwara et al., Figure 8: Schematic illustration of the terrain model used in the analysis. Crust is modeled as a set of rectangular prism-shaped bodies. Size of a color tile indicates size of the magnetic prism. The white box shows the observation area in calculation. Figure 9: Magnetic anomaly due to the terrain model. Contour interval is 50 nt. Thick black lines delineate bathymetry of 200 m contours. 37

8 Magnetic and Bathymetric Survey of the Suiyo Cross-Chain, Izu-Bonin Arc the magnetization each other out. Or the weak magnetization is caused by a combination of these. The observed anomaly above the SVR is rather comparable in amplitude and location of the pair of positive and negative anomalies to the calculated anomaly. Thus, the assumed magnetization intensity and direction may be suitable. The assumed magnetization intensity and direction cannot reproduce the observed magnetic anomaly above the SV. The result suggests that body magnetization intensity of the SV and/or magnetization contrast with adjacent crust is stronger than 5 A/m. It may indicate the SV mainly consists of younger basaltic rocks. The magnetization intensities increase from the WSVC to the SV, i.e. from back-arc to volcanic front. The negative anomaly over the eastern flank of SV cannot be explained by positive magnetization, therefore, the eastern flank may be magnetized in negative. Otherwise, this anomaly pattern above the SV, which is positive in the western flank and negative in the eastern flank, may be interpreted as the magnetization of eastward declination. The interpretation has already suggested by Yamazaki et al. [1991]. The deflection of the magnetic declination can be explained by the clockwise rotation of the seamount together with the Philippine Sea Plate since the Oligocene [e.g. Haston and Fuller, 1991]. To evaluate the magnetization structure, information of ages and magnetic properties of rock samples collected during the NT07-07 is awaited. The local anomaly in the vicinity of the caldera of the SV summit has a reverse polarity of the dipole anomaly. It means that the vicinity of the caldera has low or reverse magnetization. Low magnetization is likely due to hydrothermal activity in the caldera. Seama et al. [2005] pointed out ~800 m extent of a non-magnetized portion by using surface and deep-tow magnetic observation (KR01-15). Acknowledgements We are grateful to the officers and crew of R/V Natsushima and ROV Hyper-Dolphin for their professional work and during the cruise. We thank Drs. M. Joshima and T. Yamazaki for providing the cesium magnetometer and for instructing us on usage of the magnetometer. Part of this work is a contribution of the research program at the IFREE, JAMSTEC. References 1) Bhattacharyya, B. K., Magnetic anomalies due to prismshaped bodies with arbitrary polarization, Geophysics, 29, , ) Blakely, R. J., Potential Theory in Gravity & Magnetic Applications, Cambridge University Press, New York, 441 pp, ) Haston, R. B. and M. Fuller, Paleomagnetic data from the Philippine Sea Plate and their tectonic significance, J. Geophys. Res., 96, , ) International Association of Geomagnetism and Aeronomy (IAGA), Division V, Working Group VMOD: Geomagnetic Field Modeling, The 10th-Generation International Geomagnetic Reference Field, Geophys. J. Int., 161, , ) Ishizuka, O., R. N. Taylor, M. Yuasa, J. A. Milton, R. W. Nesbitt, K. Uto, and I. Sakamoto, Processes controlling along-arc isotopic variation of the southern Izu-Bonin arc, Geochem. Geophys. Geosyst., 8(6), Q06008, doi: /2006gc001475, ) Seama, N., A. Nishizawa, and Y. Kawada, Geophysical features of a hydrothermal system in the Suiyo Seamount, Izu- Ogasawara Arc, Western Pacific, Oceanogr. Jpn., 14, , ) Yamazaki, T., T. Ishihara, and F. Murakami, Magnetic anomalies over the Izu-Ogasawara (Bonin) Arc, Mariana Arc and Mariana Trough, Bull Geol. Surv. Japan, 42(12), , (Received July 31, 2007) 38